A neural-counting model incorporating refractoriness and spread of excitation. I. Application to intensity discrimination.

We consider in detail a new mathematical neural-counting model that is remarkably successful in predicting the correct detection law for pure-tone intensity discrimination, while leaving Weber's law intact for other commonly encountered stimuli. It incorporates, in rather simple form, two well-known effects that become more marked in the peripheral auditory system as stimulus intensity is increased: (1) the spread of excitation along the basilar membrane arising from the tuned-filter characteristics of individual primary afferent fibers and (2) the saturation of neural counts due to refractoriness. For sufficiently high values of intensity, the slope of the intensity-discrimination curve is calculated from a simplified (crude saturation) model to be 1-1/4N, where N is the number of poles associated with the tuned-filter characteristic of the individual neural channels. Since 1 less than or equal to N less than infinity, the slope of this curve is bounded by 3/4 and 1 and provides a theoretical basis for the "near miss" to Weber's law.